Identification of abducens motoneurons, accessory abducens motoneurons, and abducens internuclear neurons in the chick by retrograde transport of horseradish peroxidase

THE JOURNAL OF COMPARATIVE NEUROLOGY 259:140-149 (1987)

Identification of Abducens Motoneurons, Accessory Abducens Motoneurons, and Abducens Internuclear Neurons in the

Chick by Retrograde Transport of Horseradish Peroxidase

J.L. LABANDEIRA-GARCIA, M.J. GUERRA-SEIJAS, L..A.G. SEGADE, AND J.M. SUAREZ-NUfiEZ

Department of Anatomy, Faculty of Medicine, University of Santiago de Compostela, Santiago de Compostela, Spain

ABSTRACT The location of the motoneurons innervating the lateral rectus, pyram-

idalis, and quadratus muscles of the chick has been determined by applica- tion of horseradish peroxidase (HRP) to these muscles and their nerve branches, and internuclear neurons in the abducens nucleus have been identified by injection of HRP into the oculomotor nucleus. Quantitative results were obtained by means of a semiautomatic image analyzer. Lateral rectus motoneurons were observed only in the ipsilateral principal abducens nucleus, where they numbered 500-550, and quadratus and pyramidalis motoneurons only in the ipsilateral accessory abducens nucleus. The 325- 375 internuclear neurons that appeared in the principal abducens nucleus contralateral to the oculomotor nucleus injected with HRP were practically confined t o the rostra1 two thirds of the nucleus, where they tended to surround the lateral rectus motoneurons in dorsal or lateral positions, though a minority of interneurons also mingled with the motoneurons in the center or at the medial face of the nucleus. Most interneurons were small and elongated, but a minority of larger interneurons morphologically similar to the lateral rectus motoneurons were also distinguishable. The 100-110 quadratus motoneurons and the 45-55 pyramidalis motoneurons mingled in the accessory abducens nucleus were larger than the lateral rectus motoneurons and sent their axons into the ipsilateral abducens nerve.

Key words: lateral rectus motoneurons, abducens interneurons, accessory abducens nucleus, nictitating membrane

The organization of the motoneurons innervating the extraocular muscles of mammals has been studied in recent years by a number of researchers (Akagi, '78; Agustine et al., '81; Porter et al., '83; Gomez-Segade and Labandeira- Garcia, '83; Labandeira-Garcia et al., '81, '83). Retrograde transport of horseradish peroxidase (HRP) has been used to investigate the oculomotor nucleus of chick embryos (Hea- ton and Wayne, '83) and the motoneurons of the trochlear muscle of the duck embryo (Sohal and Holt, '781, but no study using tracers has hitherto reported the characteris- tics of the abducens nucleus of any avian species. In partic- ular, though several authors (Graybiel and Hartweig, '74;

Baker and Highstein, '75; Steiger and Buttner-Ennever, '78, '79; Spencer and Sterling, '77; Bienfang, '78; Evinger and Baker, '83; Lianger et al., '86) have found that the abducens nucleus of various mammals contains both motoneurons innervating the lateral rectus muscle and internuclear neurons projecting to the contralateral oculomotor nucleus, no such interneurons have been described in birds.

The accessory abducens nucleus was first observed in the chick embryo (van Gehuchten, 18931, and was subsequently

Accepted October 28, 1986.

01987 ALAN R. LISS, INC.

ABDUCENS NUCLEI OF THE CHICK

TABLE 1. Numbers (after correction by F), and Perikaryon Sizes and Form Factors (means k SD) of Labeled Neurons in the Principal and Accessory Abducens Nuclei'

Number F Dc(pm) Dmax(pm) Ap(pm2) Vp(pm3) ff

Lateral rectus 500-550 0.78 27.3 _+ 3.6 31.3 k 4.7 594.9 F 158.0 11,200 i 4,482 0.78 i 0.07 Internuclear neurons 325-375 0.83

Small soma 70-80% 17.0 k 2.4 21.3 i 4.6 233.7 +_ 66.5 2,767 i 1,202 0.69 i 0.08 Large soma 20-30% 25.4 i 2.7 29.7 k 4.1 514.6 i 114.6 8,935 3,101 0.76 0.06

Quadratus 100-110 0.74 29.7 ? 3.0 37.7 rt 5.6 697.4 138.2 14,054 2 4,075 0.70 ? 0.09 Pyramidalis 45-55 0.74 31.0 _+ 3.7 35.1 i 7.0 765.8 k 182.7 16,268 rt 5,834 0.70 k 0.07

'For abbreviations and details see Materials and Methods.

141

attributed the innervation of the muscles moving the avian nictitating membrane (the pyramidalis and quadratus muscles) (Terni, '22a,b; Wedin, '53) and of the mammalian retractor bulbi muscle, which in retracting the eyeball allows the nictitating membrane to slide over the cornea (Bach-y- Rita, '71). Retrograde axonal transport of HRP has con- firmed that accessory nucleus neurons innervate the retractor bulbi of mammals (Grant et al., '79; Schnyder, '84), but retractor bulbi motoneurons have also been reported in the principal abducens nucleus (Spencer and Sterling, '77; Crandall et al., '81) and even in the oculomotor nucleus (Spencer et al., '80; Meredith et al., '81); and in the only tracer study hitherto carried out to determine the location of the motoneurons innervating the avian pyramidalis and quadratus muscles, Bravo and Inzunza ('851, who studied the species Columba livia, Milvago chimango, and Tyto alba, concluded that in birds the motor innervation of these muscles is provided solely by the oculomotor complex. How- ever, Bravo and Inzunza's findings are not supported by the results reported in the present article.

In the work described in this paper retrograde transport of HRP was used to investigate the possible presence of interneurons in the principal abducens nucleus of the chick and t o locate the motoneurons innervating the lateral rectus, quadratus, and pyramidalis muscles of the same bird.

MATERIALS AND METHODS Twenty young chicks (Gallus domesticus) weighing 500-

750 g were anesthetized by intramuscular injection of 30- 35 mgkg of Nembutal and supplementary doses of ethyl ether. In the first series of experiments the lateral rectus, quadratus, or pyramidalis muscle was dissected (care being taken not to damage any nerve entering the muscle), the insertions of neighboring extraocular muscles on the sclera were cut, and the wounded ends of these muscles were cauterized. The muscle to be injected was then completely isolated from its neighbors by cottonwool pads and over a period of 30-40 min was injected with 5-10 p1 of 30% HRP (Serva) in 2% dimethylsulfoxide with a 5-p1 Hamilton syringe, the muscle being pierced at just one or two points and the needle being introduced progressively so as to inject numerous small quantities of tracer at various points throughout the muscle. To guard further against diffusion of the enzyme to neighboring muscles the cottonwool pads were renewed several times during the injection process and again when injection was complete. In a second series of experiments the nerve branch innervating the pyramidalis or quadratus muscle was cut just before its entry in the muscle, and the proximal wounded end was isolated with parafilm and cottonwool and left for 1 h in contact with HRP crystals. In a third series, the oculomotor nucleus was injected stereotaxically with 0.03-0.1 pl of 30% HRP

in 2% dimethylsulfoxide over a period of 45-60 min by means of a 1-pl HamiIton syringe or a glass micropipette connected to a WPI nanolitre pump.

After a survival time of 30-40 h the chicks were anesthetized again, heparinized, and perfused by transcardiac bi- carotid catheterization over a period of 30 min first with isotonic saline and then with 800-1,000 ml of 2.5% glutar- aldehyde in 0.1 M phosphate buffer at pH 7.2. The brain was removed from the skull, the brainstem was blocked transversely, and the blocks were postfixed for 4-6 h in the same fixative as before and then washed for 6-12 h in 0.1 M phosphate buffer at pH 7.2. Transverse serial sections 75 pm thick were cut with a vibrotome and collected in phosphate buffer, and alternate sections were developed with tetramethylbenzidine (TMB) (Gomez-Segade et al., '80; Me- sulam, '82) and o-tolidine (dimethylbemidine) (Segade, '87). Sections treated with TMB were left out to dry, cleared in xylene, and mounted without counterstaining, whereas those treated with o-tolidine were lightly counterstained with pyronin Y.

In sections treated with o-tolidine and pyronin Y, those cells with clearly visible nucleoli were measured with a Kontron Videoplan semiautomatic image analyzer. The parameters determined included the maximum diameter of the perikaryon (Dmax), the area of the perikaryon section (Ap), the diameter of the circle of area Ap (Dc), the volume of the sphere of diameter Dc (Vp), and the form factor of the perimeter [ff = 47r Ap/(perimeter)'], which approaches un- ity as the shape of the perikaryon section becomes rounder. In counting neurons, overcounts due to perikarya appearing in more than one section were corrected for by multiply- ing the number counted by the Konigsmark's correction factor: F = t/t + 2[?-(k/2p]", where t is the thickness of the sections (75 pm), r the mean value of the radii of circles with the same area as the perikarya counted, and k the diameter below which the perikarya are not measured (Konigsmark, '70). The neuron population counted and the parameters measured are listed in Table 1.

RESULTS Principal abducens nucleus

The principal abducens nucleus, whose rostrocaudal length was about 1,500 pm, was located in the rostra1 rne- dulla oblongata immediately lateral to the fasciculus longitudinalis medialis (FLM) and ventral to the fourth ventricle. The dorsalmost neurons of the rostra1 third of the nucleus lay immediately ventral to the ventricle, but the middle and caudal thirds, the latter especially, were separated from the ventricle by fascicles of fibers, most of which probably originated in or terminated at the vestibular or cochlear nuclei. These fascicles separated the abducens nucleus dorsolaterally from the nucleus vestibularis medialis.

142 J.L. LABANDEIRA-GARCIA ET AL.

Abbreviations

Axons of pyramidalis and quadratus motoneurons Principal abducens nucleus Aqueductus cerebri Corpus trapezoideum Nucleus oculomotorius, pars dorsolateralis Nucleus oculomotorius, pars dorsomedialis Nucleus of Edinger-Westphal Fasciculus longitudinalis medialis Lemniscus spinalis Nucleus olivaris superior Nucleus paramedianus Nuclei raphes Nucleus reticularis gigantocelularis Nucleus reticularis parvocelularis Nucleus tangentialis Nucleus et tractus descendens nervi trigemini Ventricle Nucleus vestibularis medialis Nervus octavus Nucleus oculomotorius, pars ventromedialis

aa AB AQ CTz DL DM EW FLM LS 0s PaM R Rgc RPC Ta mD V Vem VIII VM

Ventral and ventrolateral to the abducens nucleus lay the nucleus reticularis pontis caudalis, through which the roots of the abducens nerve passed before crossing through the corpus trapezoideum to leave the brainstem. Rather than being composed solely of large multipolar cells, the nucleus exhibited cells with differing perikaryon morphology and different roles: when injected in the lateral rectus muscle, HRP labeled numerous large, multipolar motoneurons in the ipsilateral abducens nucleus; when the tracer was injected in the oculomotor nucleus (Fig. l), a rather smaller number of labeled interneurons appeared in the contralat- era1 abducens nucleus (Fig. 2).

Lateral rectus motoneurons Injection of lateral rectus muscle labeled 500-550 neu-

rons spread over 1,425-1,500 pm from the caudal tip of the nucleus to close to the rostral tip. Some 40-50% lay in the middle third of the nucleus, as against 25-30% in each of the other two thirds. In the rostral two thirds there were more motoneurons in the ventral region of the nucleus than in the dorsal region, and more in the medial area adjacent to the FLM than in the lateral area (Fig. 3A). In the caudal third they occupied the entire section of the nucleus.

In no experiment were labeled neurons found in the contralateral abducens nucleus, the oculomotor nucleus, or the accessory abducens nucleus. On the other hand, all the animals examined exhibited 4-10 labeled neurons close to the roots of the abducens nerve at some distance from the abducens nucleus.

The labeled lateral rectus motoneurons were large (594 * 158 pm2 in area) and multipolar, and their axons, which were particularly visible under darkfield when stained with TMB, ran ventrolaterally to exit from the brainstem in the sixth cranial nerve.

Internuclear neurons When HRP was injected into the oculomotor nucleus,

325-375 neurons were labeled in the contralateral abducens nucleus, almost all of them in the rostral and middle thirds (4843% and 46-51%, respectively) and hardly any (1-2%) in the caudal third. The rostral tip of the nucleus was made up exclusively of these interneurons, which ap-

Fig. 1. Photomicrograph showing the injection site in the left oculomotor nucleus corresponding to the interneurons represented in Figure 2. The boundary of the right. oculomotor nucleus is shown by a dashed line and the right FLM by a dotted line. A dense, uniform deposit of the reaction product can be observed in the dorsomedial and vetromedial subnuclei (the former the emplacement of medial rectus motoneurons), while the dorsolateral subnucleus and the Edinger-Wesphal nucleus are less intensely stained. Strongly labeled fibers can he seen in the FLM and in both oculomotor nerves (small arrows), those of the right nerve presumably running from the superior rectus motoneurons located in the heavily stained left ventro- medial subnucleus. No labeled neurons were found in the left abducens nucleus. o-Tolidine method. ~ 3 0 .

peared 75-100 pin before the rostralmost motoneurons, in a dorsal position close to the floor of the fourth ventricle. At more caudal levels, where motoneurons were also present, the interneurons were located mainly in the dorsal and lateral areas of the nucleus, though smaller numbers also lay along the medial face or intermingled with the motoneurons in the central region. The numbers of interneurons began to fall off some 150-200 pm rostral to the caudal third, in which only a few were present on the ventrolateral or medial faces of the nucleus. Sections of the rostral two thirds of the abdiucens nucleus thus had the general shape of a triangle with a ventral base and a dorsal apex. The interneurons tended to surround the motoneurons, especially dorsally and laterally, though some mingled with the motoneurons in the central area (Fig. 3B). In the caudal third, where very few interneurons were present, the cross section of the nucleus diminished accordingly and its shape became that of an ellipse whose major axis lay dorso- ventrally.

On the basis of their size, the interneurons could be divided in two subpopulations whose members also differed somewhat in their perikaryon morphology, the large interneurons (Ap = 514 * 114 pm2) being rounder than the

ABDUCENS NUCLEI OF THE CHICK 143

Fig. 2. Camera lucida drawings showing the location of labeled neurons in the right principal abducens nucleus of the chick after injection of HRP into the right lateral rectus muscle (circles) and the left oculomotor nucleus (squares). The representation of motoneurons is taken from one animal and that of interneurons from another, and in both cases the drawings correspond to single 75-pm sections taken every 150 pm from the rostra1 tip (1) to the caudal end (10) of the nucleus.

small ones (Ap = 233 + 66 pm2, ff = 0.69 f 0.08), which made up 70-80% of the total interneuron population. Though the large interneurons were on average slightly smaller than the motoneurons (Table l), it was impossible to distinguish them individually from the latter by their shape and sue alone. Neither of the two subgroups of the interneurons exhibited any tendency to occupy any partic- ular region of the nucleus.

No internuclear neurons were observed in the abducens nucleus on the same side as the injected oculomotor nucleus. On the contralateral side, a few labeled cells generally appeared outside the abducens nucleus, either among the fibers of the FLM or ventral or ventrolateral to the nucleus (Fig. 2 , drawings 3 and 6). Numerous neurons were also of course labeled in the vestibular nuclei, specifically in the nucleus vestibularis superior (bilaterally, but mainly

144 J.L. LABANDELRA-GARCIA ET AL.

Fig. 3. Lightfield photomicrographs of transverse sections through the middle third of the principal abducens nucleus (at about level 5-6 of Figure 2), after injection of HRP into the ipsilateral lateral rectus muscle (A) and the contralateral oculomotor nucleus (B). The labeled motoneurons tend to

occupy the central area of the nucleus (A), whereas the labeled interneurons (arrows) tend to surround the central area (B). Unlabelled neurons lightly stained with pyronin Y accordingly appear mainly at the periphery in (A) and in the center in (B). o-Tolidine method. X 116.

Fig. 4. Lightfield photomicrographs of transverse sections through the fasciculi longitudinales mediales a t the level of the principal abducens nucleus after injection of HRP in the left oculomotor nucleus. Axons of vestibular and internuclear neurons can be seen before their junction with the heavily labeled left FLM. The internuclear axons (A) circumvent the dorsal boundary of the unlabeled right FLM (small arrow), or (B) cross through it more ventrally (large arrow). o-Tolidine method. A x 113; B, x 126.


Vlll

Fig. 5. Camera lucida drawings showing the location of the labeled neurons in the right accessory abducens nucleus after injection of HRP into the right quadratus muscle (circles) and the right pyramidalis muscle (squares). The representation for each muscle has been taken from a different animal, and in each case the drawings correspond to single 75-pm sections taken every 150 pm from the rostra1 tip (1) to the caudal end (8) of the nucleus.

ipsilaterally) and in the nucleus tangentialis and nucleus fascicles described above but was also labeled in its ventro- vestibularis medialis (both mainly contralaterally). How- lateral region, far from the abducens nucleus. ever, there was no risk of confounding any of these vestib- The FLM on the same side as the injected oculomotor ular neurons with the internuclear neurons of the abducens nucleus was intensely labeled, and the axons belonging to nucleus: the closest vestibular nucleus (the vestibularis me- vestibular neurons and internuclear abducens neurons were dialis) was not only separated from the interneurons by the clearly visible, so that it was possible to trace the course


taken by a number of them through several consecutive sections. Some of the interneuron axons circumvented the ipsilateral FLM at the latter's dorsal face to join the contralateral FLM, while others crossed through the ipsilateral bundle more ventrally (Fig. 4).

Accessory abducens nucleus In transverse sections the accessory abducens nucleus

was located between the nucleus olivaris superior and the nucleus descendens nervi trigemini; it consisted of 150-160 large multipolar cells spread rostrocaudally over 1,200- 1,350 pm, the rostral tip lying approximately level with the middle of the principal abducens nucleus. When HRP was injected in the quadratus muscle or applied as crystals to the corresponding nerve, 100-110 motoneurons were labeled throughout the ipsilateral accessory abducens nucleus, about 50% of them in the middle third. On average, these motoneurons were larger (Ap = 697 138 pm2) and more elongated (ff = 0.70 k 0.09) than those of the principal abducens nucleus. When HRP was injected into the pyramidalis muscle or applied to its nerve, 45-55 heavily labeled neurons appeared spread over 1,125-1,200 pm in the ipsilateral accessory nucleus, about 50% of them in the middle third. These neurons were similar to those of the quadratus muscle in being larger and more elongated than those of the principal abducens nucleus. The quadratus and pyramidalis motoneurons intermingled throughout the accessory nucleus without any tendency to form subnuclei (Figs. 5 and 6A,C).

Both when HRP was injected into the quadratus muscle or pyramidalis muscle and when it was applied directly to the nerve, the only nucleus in which labeled neurons were found was the accessory abducens nucleus, neither the principal abducens nucleus nor the oculomotor nucleus being affected. In most animals, however, between one and three labeled neurons were observed between principal and accessory abducens nuclei near the axons sent out by the latter. These axons, whose course was easily followed under darkfield in sections stained with TMB (Fig. 6B,D) could be observed in all sections featuring the roots of the abducens nerve, the distal and middle segments of most of these axons appearing in sections rostral to the rostral tip of the accessory nucleus. On leaving their perikarya, the axons headed in a dorsomedial rostral direction towards the principal abducens nucleus and then turned to continue on a ventrolateral heading with the axons of the lateral rectus motoneurons. Most of the accessory nucleus axons changed direction on or before arriving at the ventral face of the principal abducens nucleus, but some penetrated the nucleus before turning, a few even reaching its dorsolateral region.

DISCUSSION The survival of most species of birds depends heavily on

their visual system, which is generally highly developed. Compared with that of the head, the volume of the eye is much greater than in mammals, and in some birds the combined weight of the two eyes is even greater than that of the brain. For this reason birds have been extensively used in research on vision, notwithstanding which there is very little information available concerning the neurology of ocular movements in birds.

Recent investigations (Steiger and Buttner-Ennever, '78.

'86) have shown that the abducens nucleus of various species of mammal contains internuclear neurons projecting to the contralateral oculomotor nucleus, specifically to the area where the motoneurons of the medial rectus muscle are located (Biittner-Ennever and Akert, '81; Furuya and Markham, '811, and that these neurons play an essential role in the coordination of the lateral rectus and medial rectus muscles during conjugate horizontal eye movements. In birds, the movements of the two eyes are often indepen- dent, but they ,are coordinated when binocular vision is involved (Kare and Rogers, '76). Most birds are capable of binocular vision, many of them following moving objects with binocular vision after first moving the entire head. Depth perceptioin is also usually achieved by binocularity. The portion of the visual field that is binocular is determined by the shape of the head, the shape of the eye, and the position of the eye in the head. Birds with flat, laterally situated eyes, such as chickens and pigeons, have a rela- tively smaller binocular field than those with tubular eyes located frontally, such as owls and hawks. Whatever the size of the binocular field, however, the existence of coordinated movemenits of the two eyes during binocular vision makes it reasonable to hypothesize that in birds, as in mammals, the abducens nucleus must be connected, directly or indirectly, with the oculomotor complex.

Isomura ('73) reported that the abducens nucleus of the chick consisted of 162-261 large multipolar neurons. This homogeneous neuron population contrasted with those of the oculomotor and trochlear nuclei, which contained both large and small multipolar cells. Only 8-19% of the abducens nucleus cells were chromatolytic after transection of the ipsilateral albducens nerve. In the present study the number of motoneurons observed to innervate the lateral rectus from the abducens nucleus was much greater; and it was found that, ,as in mammals, the abducens nucleus was directly connected to the contralateral oculomotor nucleus by internuclear neurons that were located in the former and whose axons reached the latter via the contralateral FLM. Thus in the chick, as in mammals, there are major afferents to the oculomotor complex not only from the vestibular complex but also from the internuclear contralat- era1 abducens neurons.

In the material used in this study we observed no diffusion of the enzyme from the oculomotor nucleus to other mesencephalic structures in which the axons of internuclear abducens neurons might terminate. Since the oculomotor nucleus is separated from surrounding structures by the FLM, it is possible that there may have been some uptake of HRP by some of the FLM axons closest to the injection site, but this circumstance, if the fact existed, seems unlikely to have any effect on our results: For in first place, uptake and transport of HRP by intact axons of passage seems to be negligible in the central nervous system (Mesulam, '82); and second, it seems reasonable to assume (even though it has not been demostrated in birds by tracer studies) that among the FLM axons close to the oculomotor nucleus, those terminating in the oculomotor nucleus itself greatly outnumber those terminating in more rostral nuclei.

The internuclear neuron group found in the abducens nucleus of the chick presents various similarities with those reported in mammals. As in the cat (Steiger and Biittner- Ennever, '78; Highstein et al., '82). the interneurons lie

'79; Glicksman, %0; Evingerand Baker, '83; Langer et al.; mainly in the r'okral two thirds of the nucleus; and al-


Fig. 6. Photomicrographs of the accessory abducens nucleus, A,B: La- beled quadratus muscle motoneurons in the rostra1 third of the nucleus surrounding the nucleus olivaris superior. C, D: Labeled pyramidalis muscle motoneurons in the more rounded caudal third of the nucleus. Labeled axons (large arrows) are visible in the darkfield photomicrographs of TMB- stained sections (B,D), in which the most heavily labeled perikarya are seen

only as outlines (small arrows). In the lightfield photomicrographs of 0- tolidine-stained sections (A,C), motoneurons heavily labeled with HRP are intermingled with motoneurons of the uninjected muscle (the pyramidalis in A and the quadratus m C), which are lightly stained with pyronin Y. A, x221; B, x 126; C, X 178; D, X 126.


(very weakly lilbeled) in the dorsolateral and dorsomedial subnuclei of the oculomotor nucleus and the dorsal trigem- inal motor nucleus was in an experiment in which HRP was seen to escape from the pyramidalis muscle during injection and t~he labeling of the latter nuclei was accordingly attributed to tracer's diffusion to the muscles nearest the pyramidalis (the inferior rectus, medial rectus, and depressor palpebrae inferioris). We suggest the possibility that, although the difference in species should of course be taken into account, the label neurons observed by Bravo and Inzunza ('85) in the oculomotor nucleus may have been the result of tracer's diffusion to other extraocular muscles.

In cats (Spencer et al., '80; Baker et al., '80) and in rabbits (Gray et al., '811, retractor bulbi motoneurons have been reported not only in the accessory abducens nucleus but also in the principal abducens nucleus and even in the oculomotor nucleus. However, only the accessory abducens nucleus is considered responsible for defensive globe retraction and hence for the nictitating membrane's covering the cornea (Baker et al., '80; Meredith et al., '81; Berthier and Moore, '83). The retractor bulbi motoneurons observed in the oculomotor and principal abducens nuclei are regarded as responsible for patterned (i.e., rotational) eye movements. If this is so, the absence of pyramidalis and quadratus motoneulrons from the oculomotor and principal abducens nuclei of the chick is not surprising, since these muscles act directly on the nictitating membrane without displacing the eyeball, and they seem poorly situated in the orbit for any involvement in rotational eye movements.

though there is considerable overlap between the interneurons and motoneurons in the cat nucleus (Spencer and Sterling, '77), a certain tendency to cluster on the bounda- ries of the nucleus as in the chick has also been observed in this animal (Steiger and Buttner-Ennever, '78; Highstein et al., '82). In the rat (Glickman, '80), guinea pig (Evinger and Baker, '831, and monkey &anger et al., '86) the interneurons are likewise located mainly in lateral or ventrolat- era1 positions. On the whole, current data seem to suggest that in animals with laterally situated eyes, such as the rat and the chick, the interneurons are rather more segregated from the motoneurons than in animals with more frontal eyes and wider binocular visual field, such as cats and monkeys. The presence of two different types of interneuron, one small and fusiform and other larger and morphologically similar to the motoneurons, has also been observed in the cat (Spencer and Sterling, '77; Highstein et al., '82). Finally, the path followed by the interneurons' axons towards the contralateral FLM is similar to that which has been elegantly demonstrated in the cat (Highstein et al., '82) and in the monkey (McCrea et al., '86) by means of intracellular injection of HRP.

In mammals, nictitating membrane extension occurs when the retractor bulbi muscle executes reflex retraction of the eyeball within the orbit, thus allowing the nictitating membrane to slide over the cornea, being both protective and lubricating (Bach-y-Rita, '71). Avian nictitating membrane extension is brought about by a different mechanism, in which the nictitating membrane is moved by the pyramidalis and quadratus muscles. Both these muscles are known to have derived from the retractor bulbi of reptilian ancestors via a series of intermediate structures found in reptiles and amphibians (Rochon-Devigneaud, '54). The action of the pyramidalis and quadratus muscles is transmit- ted by a slender tendon that originates at the nictitating membrane, borders the lower eyeball, and turns in dorso- medially to make a bight over the optic nerve and attach to the pyramidalis. In crossing over the optic nerve, the tendon passes through a pulley formed by the ventral extremity of the quadratus, whose other extremity is inserted on the dorsal caudal quadrant of the sclera. This singular muscular arrangement allows the action of the pyramidalis to be modulated by the quadratus, and apparently protects the optic nerve from the pressure of the tendon.

In the past, various authors have attributed to the accessory abducens nucleus of birds the innervation of the quadratus and pyramidalis muscles (Terni, '22a,b; Wedin, '53). However, after injecting HRP into these muscles in three different species (Columba livia, Milvago chimango, and T'to alba), Bravo and Inzunza ('85) concluded that in birds they are innervated solely by the oculomotor nucleus, hav- ing found no labeled neurons in the principal abducens nucleus (no mention was made of the accessory abducens nucleus). In the experiments described in the present article we always found that heavily labeled neurons appeared in the accessory abducens nucleus after HRP had been injected into the quadratus or pyramidalis muscles. These muscles were innervated by slender ramifications from a narrow branch that derived from the abducens nerve, and when HRP crystals were applied directly to the nerve branches entering the muscles, labeled neurons again appeared exclusively in the accessory abducens nucleus. The onlv occasion on which labeled neurons anaeared not onlv

LITERATURE CITED Agustine, J.R., E.G. Deschamps, and J.G. Ferguson, Jr. (1981) Functional

organization of the oculomotor nucleus in the baboon. Am. J. Anat. 161t393-403.

Akagi, Y. (1978) The localization of the motor neurons innervating the extraocular muscles in the oculomotor nuclei of the cat and rabbit, using horseradish peroxidase. J. Comp. Neural. 181:745-762.

Baeh-y-Rita, P. (1971) Neurophysiology of eye movements. In P. Bach-y-Rita and C. Collins (ed): The Control of Eye Movements. New York: Aca- demic, pp. 7-45.

Baker, R., and S.M. Highstein (1975) Physiological identification of interneurons and motoneurons in the abducens nucleus. Brain Res. 83292- 298.

Baker, R., R.A. McCrea, and R.F. Spencer (1980) Synaptic organization of cat accessory abducens nucleus. J. Neurophysiol. 43:771-791.

Berthier, N.E., and J.W. Moore (1983) The nictitating membrane response: An electrophysiological study of the abducens nerve and nucleus and the accessory abducens nucleus in rabbit. Brain Res. 258:201-210.

Bienfang, D.C. (1978) The course of direct projections from the abducens nucleus to the contralateral medial rectus subdivision of the oculomotor nucleus in the cak. Brain Res. I45:277-289.

Bravo, H., and 0. Inzunza (1985) The oculomotor nucleus, not the abducent, innervates the muscles which advance the nictitating membrane in birds. Acta Anat. 122r99-104.

Buttner-Ennever, J.A., and K. Akert (1981) Medial rectus subgroups of the oculomotor nucleus and their abducens internuclear input in the monkey. J. Comp. Neural. I97t17-27.

Crandall, W.F., S. J. (Goldberg, J.S. Wilson, and J.R. McClung (1981) Muscle units divided among retractor bulbi muscle slips and between the lateral rectus and rsetractor bulbi muscles in cat. Exp. Neurol. 71951-260.

Evinger, C., and R. Baker (1983) Localization and morphology of guinea-pig extraocular motoneurons and internuclear neurons. Soc. Neurosci. Abstr. 9:1088.

Furuya, N., and C.H. Markham (1981) Arborization of axons in oculomotor nucleus identified by vestibular stimulation and intra-axonal injection of horseradish peroxidase. Exp. Brain Res. 43:289-303.

Glicksman. M.A. (1980) Localization of motoneurons controlling the extra- in ihe accessory abducens nucleus (heavily iabeled) but als'b ocular muscles of rat. Brain Res. 188:53-62.

ABDUCENS NUCLEI OF THE CHICK

Gomez-Segade, L.A., and J.L. Labandeira-Garcia (1983) Location and quantitative analysis of the motoneurons innervating the extraocular muscles of the guinea-pig, using horseradish peroxidase (HRP) and double or triple labelling with fluorescent substances. J. Hirnforsch. 24.613- 626.

Gomez-Segade, L.A., J.L. Labandeira-Garcia, and J.M. Suarez-Nuiiez (1980) Some considerations pertaining to the use of horseradish peroxidase (HRP) in neuroanatomy. Trab. Inst. Cajal 71:5-23.

Grant, K., J.P. Guetiraud, G. Horcholle-Bossavit, and S. Tye-Dumont (1979) Anatomical and electrophysiological identification of motoneurones supplying the cat retractor bulbi muscle. Exp. Brain Res. 34r541-550.

Gray, T.S., S.E. McMaster, J.A. Harvey, and I. Gomerzano (1981) Localiza- tion of retractor bulbi motoneurons in the rabbit. Brain Res. 226:93- 106.

Graybiel, A.M., and E.A. Hartweig (1974) Some afferent connections of the oculomotor complex in the cat: An experimental study with t,racer tech- niques. Brain Res. 81543-551.

Heaton, M.B., and D.B. Wayne (1983) Patterns of extraocular innervation by the oculomotor complex in the chick. J. Comp. Neurol. 216:245-252.

Highstein, S.M., A. Karabelas, R. Baker, and R.A. McCrea (1982) Compari- son of the morphology of physiologically identified abducens motor and internuclear neurons in the cat: A light microscopic study employing the intracellular injection of horseradish peroxidase. J. Comp. Neurol. 208r369-381.

Isomura, G. (1973) On the localization of nerve centers of the extrinsic ocular muscles in the fowl. Anat. Am. Bd. I33:35-50.

Kare, M.R., and J.G. Rogers (1976) Sense organs. In P.D. Sturkie (ed): Avian Physiology. BerlidNew York: Springer-Verlag, pp. 28-52.

Konigsmark, B.W. (1970) Methods for the counting of neurons. In W.J.H. Nauta and S.O.E. Ebbesson (eds): Contemporary Research Methods in Neuroanatomy. BerlidNew York: Springer-Verlag, pp. 313-340.

Labandeira-Garcia, J.L., L.A. Gomez-Segade, and J.M. Suarez-Nutiez (1981) Organizacih de las motoneuronas en 10s nucleos del tercero, cuarto y sexto pares craneales en el conejo. An. Anat. 30:347-366.

Labandeira-Garcia, J.L., L.A. Gomez-Segade, and J.M. Snarez-Nufiez (1983) Localisation of motoneurons supplying the extra-ocular muscles of the rat using horseradish peroxidase and fluorescent double labelling. J. Anat. 137:247-261.

Langer, T., C.R.S. Kaneko, C.A. Scudder, and A.F. Fuchs (1986) Afferents to the ahducens nucleus in the monkey and cat. J. Comp. Neurol. 189:191- 209.

McCrea, R.A., A. Stassman, and S.M. Highstein (1986) Morphology and physiology of abducens motoneurons and internuclear neurons intracel- lularly injected with horseradish peroxidase in alert squirrel monkeys. J. Comp. Neurol. 213:291-308.

Meredith, M.A., J.R. McClung, and S.J. Goldherg (1981) Retractor bulbi muscle responses to oculomotor nerve and nucleus stimulation in the cat. Brain Res. 211t427-432.

Mesulam, M.M. (1982) Principles of horseradish peroxidase neurohistochemistry and their applications for tracing neural pathways-axonal transport, enzyme histochemistry and light microscopic analysis. In M.M. Mesulam (ed): Tracing Neural Connections with Horseradish Per- oxidase. Chichesterrnew York Wiley, pp. 1-153.

Porter, J.D., B.L. Guthrie, and D.L. Sparks (1983) Innervation of monkey extraocular muscles: Localization of sensory and motor neurons by retrograde transport of horseradish peroxidase. J. Comp. Neurol. 218r208- 219.

Rochon-Duvigneaud, A. (1954) L’oeil des vertebres. In P.P. Grasse (ed): Traite de Zoologie. Paris: Masson, Vol. 12, pp. 333-452.

Schnyder, H. (1984) The innervation of the monkey accessory lateral rectus muscle. Brain Res. 296139-144,

Segade, L.A.G. (1987) Pyrocatechol as a stabilizing agent for o-tolidine and o-dianisidine: A sensitive new method for HRP neurohistochemistry. J. Hirnforsch. (in press).

Sohal, G.S., and R.K. Holt (1978) Identification of the trochlear motoneurons by retrograde transport of horseradish peroxidase. Exp. Neurol. 59509- 514.

Spencer, R.F., and P. Sterling (1977) An electron microscope study of motoneurons and interneurons in the cat abducens identified by retrograde intraaxonal transport of horseradish peroxidase. J. Comp. Neurol. 176%-86.

Spencer, R.F., R. Baker, and R.A. McCrea (1980) Localization and morphology of cat retractor bulbi motoneurons. J. Neurophysiol. 43:754-770.

Steiger, H.J., and Buttner-Ennever (1978) Relationship between motoneurons and interneurons in the abducens nucleus: A double retrograde tracer study in the cat. Brain Res. 148:181-188.

Steiger, H.J., and J.A. Buttner-Ennever (1979) Oculomotor nucleus afferents in the monkey demonstrated with horseradish peroxidase. Brain Res. 16O:l-15.

Terni, T. (1922a) Richerche sul nervo abducente e in special mod0 intorno a1 significato del suo nucleo accessorio d‘origine. Folia Neurobiol. 12277- 327.

Terni, T. (192213) I1 sostrato anatomic0 del riflesso di chiusura dell mem- brana nittitante nei sauropsidi. Arch. Fisiol. 20:305-311.

Van Gehuchten, A. (1893) Anatomie du Systeme Nerveux de 1’Homme. Louvain: Uyspruyt, p. 383.

Wedin, B. (1953) The development of the eye muscles in Ardea cinerea. Acta Anat. 18:30-48.

Documents

Identification of abducens motoneurons, accessory abducens motoneurons, and abducens internuclear neurons in the chick by retrograde transport of horseradish peroxidase